Journal of Environmental Radioactivity 150 (2015) 57e61

Contents lists available at ScienceDirect

Journal of Environmental Radioactivity

journal homepage: www.elsevier.com/locate/jenvrad

New study on the correlation between carbon dioxide concentration

in the environment and radon monitor devices
bia
n a, T. Kova
cs a
A. Shahrokhi a, *, B.D. Burghele b, F. Fa
a
b

Doctoral School of Chemistry and Environmental Sciences, Institute of Radiochemistry and Radioecology, University of Pannonia, Veszpr
em, Hungary
Interdisciplinary Research Institute on Bio-Nano-Sciences of Babes-Bolyai University, Cluj-Napoca, Romania

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 24 April 2015
Received in revised form
30 June 2015
Accepted 26 July 2015
Available online xxx

The inuence of high geogenic carbon dioxide concentrations on monitoring devices might present a
signicant challenge to the measurement of radon concentrations in environments with a high level of
carbon dioxide concentration such as volcano sites, mofettes, caves, etc. In this study, the inuence of
carbon dioxide concentration on several different types of radon monitor devices d including Alpha
Spectrometry (Sarad RTM 2200, EQF 3220, RAD7), Ionizing Chamber (AlphaGUARD PQ2000 PRO) and
Active Cell (Active scintillation cell, Pylon 300A) d was examined to represent new aspects of radon
measuring in environments with carbon dioxide. In light of the results, all measuring devices were
exposed to variable conditions affected by carbon dioxide concentration, except for the AlphaGUARD,
which was kept in a steady state throughout the experiment. It was observed that alpha spectroscopy
devices were affected by carbon dioxide, since measured radon concentrations decreased in the presence
of 70% and 90% carbon dioxide concentrations by 26.5 2% and 14.5 2.5% for EQF 3220, and 32 2%
and 35.5 2% for RTM 2200. However, the ionizing chamber instrument was unaffected by changes in
carbon dioxide concentration. It was determined that the RAD7 performed relatively inefciently in the
presence of carbon dioxide concentrations higher than 67% by an overall efciency factor of approximately 0.52, conrming that it is not an admissible radon monitor instrument in environments with high
carbon dioxide concentrations.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Radon
Carbon dioxide
RAD7
AlphaGUARD
Sarad
Pylon
Lucas cell

1. Introduction
Radon (Rn-222) emanation measurements are frequently performed in geological and environmental studies where high concentrations of carbon dioxide are often found in soil, caves, mofettes
(natural CO2 spring areas) or fumaroles adjacent to volcanoes
(Bonforte et al., 2013; Zimmer and Erzinger, 2003; Phuong et al.,
rez et al., 2007; Papp et al., 2010).
2012; Heiligmann et al., 1997; Pe
In this context, it should be mentioned that in the presence of carrier
gases such as methane (CH4) or carbon dioxide (CO2), the radon ux
might be larger than in other environments (Martini et al., 2010;
Frunzeti et al., 2013). The theory of this phenomenon states that
deep, magmatic sources are characterized by high radon activities
and high CO2 efuxes, while shallow sources are indicated by high
thoron activities and rather low CO2 efuxes (Giammanco et al.,
2007). However, the importance of this theory arises from the fact

* Corresponding author.
E-mail address: ashahrokhi@almos.uni-pannon.hu (A. Shahrokhi).
http://dx.doi.org/10.1016/j.jenvrad.2015.07.028
0265-931X/ 2015 Elsevier Ltd. All rights reserved.

that certain mofette sites (natural CO2 springs), characterized by an

elevated level of CO2, can be used as monitoring, assessment and
forecasting tools in geographical studies, e.g., in monitoring faults;
to assess the potential risks of a volcano, using CO2 as a tracer; to
forecast a volcano's potential, etc. (Moloney et al., 2011; Saurer et al.,
2003). Escaping CO2 from deep underground contains an amount of
natural radioactive Rn-222 gases that it is important to monitor, not
only due to radiation dosimetry aspects, i.e., a strong correlation
between radon exposure and lung cancer occurrence (Somlai et al.,
2011), but also for geographical studies, i.e., for a better understanding of radon exhalation associated with tectonic or volcanic
processes (Richon et al., 2005). Consequently, it should be mandatory to monitor the radon levels of mofettes used for curative purposes due to high level of the radon activity concentration in the
mofettes' indoor air and dosimetery is needed to estimate the
da et al., 2008a,
effective dose received by workers and patients (Ne
2008b). Lane-Smith and Sims (2013) have recently shown that the
active Rn-220 and Rn-222 measuring devices could be affected by
the presence of CO2 while performing radon or thoron measurements. They noted that for every percentage of CO2 concentration,

58

A. Shahrokhi et al. / Journal of Environmental Radioactivity 150 (2015) 57e61

the calculated Rn-220 concentration should be multiplied by a

correction factor of 1.019, whereas the Rn-222 concentration ought
to be multiplied by 1.003, on account of Lane-Smith and Sims' observations regarding the effects that CO2 might exert on radon
measurements performed with radon monitor devices used
worldwide (LaneeSmith and Sims, 2013).
In the present study, a new experiment was carried out to
determine the inuence of variable carbon dioxide concentrations
on the most common types of radon monitor devices available on
the market.
2. Materials and methods
In order to analyse the potential effects of environmental carbon
dioxide on the radon monitor devices, a system was assembled to
facilitate the use of a controllable Rn-222 and CO2 concentration in
a conned space. A newly developed and certied 210.5 L metal
radon chamber (Genitron EV 03209) with a removable lid, and an
electric fan in order to ensure internal homogenization, was used as
an experimental chamber. The carbon dioxide was fed to the
chamber through a system of tubing connected to a ow meter and
a CO2 sensor, which continuously monitored the CO2 level inside
the chamber. The gas intake ux was kept constant, at one litre per
minute. A certied radon source (Pylon RN2000A, a passive radon
gas source) supplied a known concentration of radon through
another system of tubing within the chamber in which the experiment was conducted. The passive radon source (Pylon Electronics
Inc., Canada) used for this purpose was 105.7 0.42 kBq radium
(226Ra) covered in a leak-proof metal vessel to emanate radon gas at
a constant rate. After the criteria for a muffed atmosphere were
met, four different devices commonly used for monitoring radon
were introduced inside the above-mentioned chamber. The four
devices used in the test were chosen because they involve different
measuring methods, namely: alpha detector devices (RTM 2200
and EQF 3220, Sarad, Germany), RAD7 (Durridge Company, Inc.)
and an ionizing chamber device (AlphaGUARD PQ2000 PRO,
Saphymo Gmbh, Germany).
AlphaGUARD was set up in diffusion mode to conduct measurements over a period of 10 min, while RAD7 was set up with the
following settings: protocol User, Mode of sniff, 10 min of cycling,
unlimited recycles, Thoron off, Pump Auto and tone off. A drying
tube (CaCl2) was installed on the inlet of RAD7 to keep the humidity
inside the RAD7 container below 10%, in line with the user manual's
guide. Both EQF 3220 and RTM 2200 were set to sample for 10 min
on fast mode while these devices measured the CO2 concentration
in the air. Then all the devices were placed in the chamber, which
was sealed and vacuumed. Then the Pylon 2000A passive radon gas
source was connected to the chamber to reach a specic activity of
approximately 20 kBq/m3, whereas the radon activity concentration
inside the chamber was kept constant for the entire duration of the
experiment by use of a three-way connector Y joint with a regulator
and a ow meter connected to the outlet of the radon source
chamber and the inlet of the experiment chamber, respectively.
Meanwhile, the connector's third connection was attached to the
outside by a plastic pipe. Then, by controlling the ow rate using a
ow meter and a regulator, the radon concentration inside the
experiment chamber was kept constant with an uncertainty value of
0.2 kBq/m3. Fig. 1 presents a schema of the experimental system.
Lastly, the carbon dioxide was injected into the chamber to increase its concentration from 0% to approximately 89%. After each
sampling and measurement, the carbon dioxide concentration was
increased while the radon concentration was kept at the level
stated 20 kBq/m3. The experiment was separated into the two
phases; in the rst part, the carbon dioxide concentration was
increased from 0% up to 17%; however, in the second phase, the CO2

Fig. 1. Schematic of experiment system.

concentration inside the chamber reached 67%, and then increased

by 5% after each sampling period to 89%. In both phases, the concentration of the reference radon was about 20 kBq/m3. Simultaneously with each signicant increase in the CO2 concentration,
different radon samples were grabbed from both inside and outside
the chamber. A fth device based on grab sampling into a 272 ml
Active Cell (Pylon 300A Lucas type scintillation cell, Pylon Electronics Inc, Canada) was used to take gas samples from the experimental chamber. According to the number of counts provided by
Pylon, the radon activity concentration was later calculated using
the following Equation (1):

Rc C  B=E S 3 t V

(1)

where Rc is radon concentration (Bq/m3), C is number of counts, B is

the background count, E is efciency of the scintillation cell, is the
efciency of sampling (0.98), V is the volume of the cell
(0.27  103 m3), t is the measuring time (Second), 3 is the number
of alpha emitting radionuclides after four hours equilibrium (radon
and its two short-lived decay products), and S is a correction factor
which is calculated using S elt where l represents the decay
constant.
To maintain the activity concentration of radon inside the
chamber, the total daily radon emanation rate from the radium
source was calculated using Equations (2)e(4); then, by adjusting
the ow rate, the radon concentration remained constant during
the experiment.

REm A0 elt

(2)

l ln2=T

(3)

Rc A=V

(4)

where REm is the daily radon emanation from the source (Bq), A0 is
the initial activity that was provided by the producer (Bq), l is the
decay constant which is calculated from Equation (3), t is the time

A. Shahrokhi et al. / Journal of Environmental Radioactivity 150 (2015) 57e61

59

Fig. 2. The inuence of CO2 concentrations (>0.5%e17%) on the radon monitor devices in the presence of radon concentration at 20 kBq/m.3.

interval between the preparation of the source and using dates

(Days), T is the half-life of Ra-226 (Days), Rc is the radon concentration (Bq/m3), and V is the volume of the barrel (m3).
3. Results and discussion
The experiment was divided into the two phases of measuring
radon concentration in the presence of carbon dioxide. In the rst
aspect, radon monitor devices were exposed to CO2 concentration
of 0%e17% due to concentrations of environmental carbon dioxide.
However, in the second aspect, radon monitor devices were presented to high levels of CO2 concentrations, higher than 67% and as

high as 89%, a level more likely to occur in environments with a

high level of CO2 concentration, e.g., caves, natural CO2 spring areas
(mofettes), deep in the soil, etc.
The radon activity concentration inside the chamber was kept
constant at 20 kBq/m3 during the whole experiment. The Radium
(226Ra) source (Pylon RN2000A Solid Radium) was the reference of
radon concentration inside the chamber. However, all devices were
checked before presenting the CO2 to ensure that all can measure
the radon concentration at 20 kBq/m3, and then the carbon dioxide
was fed to the chamber. Each of the instruments was presented to
the carbon dioxide concentrations from >0.5% up to 17% at 20 kBq/
m3 activity concentration of radon during the rst phase of the

Fig. 3. A comparison between the performance of radon monitor devices exposed to low concentrations of CO2 (>0.5%e17%).

60

A. Shahrokhi et al. / Journal of Environmental Radioactivity 150 (2015) 57e61

Table 1
The response of radon monitor devices to radon concentration at 20 kBq/m3 and CO2
concentrations between 67% and 89%.
CO2 (%)

67

69

72

77

79

82

85

89

AlphaGUARD
/
EQF 3220
/
RTM 2200
/
RAD7
/
Lucas Cell
/

20.2
1
14.7
0.4
13.6
0.5
11.2
0.5
18.7
0.9

20
1
16.6
0.5
14
0.5
11
0.5
18.5
2.7

19.9
1
14.5
0.4
13
0.5
9.8
0.5
18.3
1.7

19.9
1
15.1
0.4
12.5
0.5
9.8
0.5
18.2
0.9

20.4
1
15.9
0.5
12.8
0.5
10
0.5
17.9
1.1

20.1
1
16.4
0.5
13
0.5
10
0.5
17.1
0.9

19.9
1
18.3
0.5
14
0.5
10.6
0.5
17.8
2.0

20.3
1
17.1
0.5
12.9
0.5
9.6
0.5
17.4
0.5

kBq/m3

experiment. In light of the results, all devices were inuenced by

the CO2 concentrations in analysing radon concentrations in the air,
except for the AlphaGUARD, which was not affected by variations in
carbon dioxide concentration during the experiment.
In Fig. 2, the response of each device to low concentrations of
CO2 as a function of radon concentration is shown. All devices
measured the same range of the referenced radon concentration
(20 kBq/m3) when no CO2 was present in the chamber. However,
both of Sarad's devices, EQF 3220 and RTM 2200, measured radon
concentrations with large deviations from the referenced radon
activity (20 kBq/m3 radon concentrations in the chamber) after
exposure to a CO2 concentration of up to 17%.
On the other hand, RAD7 showed no signicant deviations from
the measured radon activity concentration, apart from a slight
tendency to underestimate. Conversely, the AlphaGUARD radon
monitor remained unaffected by increasing the carbon dioxide
concentration. Fig. 3 shows the comparison and colouration of the
devices in accordance with carbon dioxide concentrations.
In the second phase of the experiment, the CO2 concentration
was raised to over 67%. In this stage, the Pylon scintillation cell and
Pylon AB-5 Portable Radiation Monitor (PRM) were also introduced
to obtain several grabbing samples with the assistance of Lucas
cells from the experimental chamber. Table 1 shows the output

results of instruments in conditions where carbon dioxide concentrations increased by up to 89%. The results show a highly
diminished radon concentration as measured by three of the devices involved.
Special attention was drawn to the RAD7 device which exhibited
a relatively constant midpoint of the reference radon activity inside
the chamber (20 kBq/m3) in conditions where CO2 concentrations
were <67%. However, RAD7 produced results much closer to the
chamber reference concentration (20 kBq/m3) in the presence of
low levels of carbon dioxide concentrations (below 17%), but when
the CO2 concentration reached 89%, RAD7 was measured as being
less than the reference radon concentration inside the chamber
(20 kBq/m3) by about 52 2%. On the other hand, RTM 2200 produced the highest deviation in terms of radon activity, with a
similar tendency manifested at low CO2 concentrations, with the
exception being that the highest activity measured using this device remains below the level of the reference activity (20 kBq/m3)
when the carbon dioxide concentration level is relatively high. In
the light of the results produced by RTM 2200, a reduction in the
measured radon was determined as 32 2.5% in the presence of a
carbon dioxide concentration of 67%, and a value of approximately
35.5 2.5% when exposed to a CO2 concentration of 89%. The EQF
3220 produced small deviations while still maintaining a lower
activity compared to the reference radon concentration level inside
the chamber. The measured radon concentrations in the chamber
were lower than the reference radon concentration by 26.5 2%
and 14.5 2.5% in the presence of carbon dioxide concentrations of
67% and 89%, respectively. The colouration of instruments exposed
to CO2 concentrations between 67% and 89% is shown in Fig. 4.
As an ionizing chamber instrument, AlphaGUARD measured the
activity concentration as being equal to the reference radon concentration inside the chamber with an uncertainty of about 2 kBq/
m3 in the presence of CO2 concentrations of less than 89%, which is
similar to its performance in conditions of low concentrations of
carbon dioxide. However, the Lucas cell was not signicantly
affected. At a CO2 concentration of 67%, the measured activity
dropped by 6.5 4.5% and continued to fall until it reached
13 2.5% at a CO2 concentration of 89%.

Fig. 4. Comparison between the performances of radon monitoring devices exposed to high concentrations of CO2 (67%e89%).

A. Shahrokhi et al. / Journal of Environmental Radioactivity 150 (2015) 57e61

61

Fig. 5. Radon monitoring devices' efciency graph as a function of the CO2 concentration (>0.5%e17% and 67%e89%).

According to Lane-Smith and Sims's research article (2013)

about the correlation between radon measurement and CO2 concentration using RAD7, a specic correction factor of 0.3% per 1%
CO2 have been suggested to be employed with the measured value,
while the present study has found that the correction factor can
change depending on the CO2 concentration in air. Fig. 5 presents
the correction factor graphs (efciency) of each device in terms of
CO2 concentration.
The humidity, temperature and pressure inside the chamber
were monitored during the experiment and were calculated as
being between 15 and 25%, 20e24  C and 987 mbar, respectively.
4. Conclusions
AlphaGUARD's performance was unaffected during the experiment, while the EQF 3220 shows unexpected behaviour in the
presence of high levels of CO2 by increasing in efciency, since the
CO2 concentrations were increased from 67% up to 89%. The test
performed using RAD7 conrms that the efciency of radon
detection decreases as the CO2 concentration increases in the carrier gas for electrostatic precipitation measuring devices. Moreover,
the experiment draws attention to the matter of other instruments,
pointing out that they could most certainly be inuenced by the
level of CO2 in the analysed air, especially when this exceeds 89% as
is the case in mofettes. Lane-Smith and Sims have suggested a
specic correction factor of 0.003 for every percentage of CO2 when
measuring radon concentration in the presence of CO2 by RAD7.
However, according to the results of this study, giving a specic
correction factor depends on the CO2 concentration in environment. It is therefore suggested that an efciency graph in terms of
CO2 concentration is provided with the manual book by companies
producing such equipment. According to the stable results of the
AlphaGUARD, it observed that the AlphaGUARD can be used as a
reference device to calibrate the other devices in the presence of
CO2 concentrations. In addition to the effects on active devices, it
should also be determined whether the level of CO2 affects measurements performed in an environment with a high level of CO2
concentration using passive methods, which seem to have been
gaining increasing interest recently.
Acknowledgement
The present study was supported by the Hungarian State and

the European Union projects Grant No. TAMOP-4.2.2.A11/1/

KONV-2012-0071 and TAMOP-4.2.2.B-15/1/KONV-2015-0004.

References
Bonforte, A., Federico, C., Giammanco, S., Guglielmino, F., Liuzzo, M., Neri, M., 2013.
Soil gases and SAR measurements reveal hidden faults on the sliding ank of
Mt. Etna (Italy). J. Volcanol. Geotherm. Res. 251 (1), 27e40.
Frunzeti, N., Moldovan, M., Burghele, B.D., Cosma, C., Papp, B., Popita, G.E.,
Stoian, L.C., 2013. Flux measurements of 222Rn, CH4 and CO2 along with soil
gas concentrations (222Rn, CO, NO2 and SO2) over a methane reservoir in
Transylvania (Romania). Carpath. J. Earth. Env. 8 (3), 75e80.
Giammanco, S., Sims, K.W.W., Neri, M., 2007. Measurements of 220Rn and 222Rn
and CO2 emissions in soil and fumarole gases on Mt. Etna volcano (Italy): implications for gas transport and shallow ground fracture. Geochem. Geophys. 8
(10), 1e14.
 n, V.,
Heiligmann, M., Stix, J., Williams-Jones, G., Sherwood Lollar, B., Gustavo Garzo
1997. Distal degassing of radon and carbon dioxide on Galeras volcano,
Colombia. J. Volcanol. Geotherm. Res. 77 (1e4), 267e283.
LaneeSmith, D., Sims, K.W.W., 2013. The effect of CO2 on the measurement of
220Rn and 222Rn with instruments utilising electrostatic precipitation. Acta
Geophys. 61 (4), 822e830.
Martini, F., Tassi, F., Vaselli, O., Del Potro, R., Martinez, M., Van del Laat, R.,
Fernandez, E., 2010. Geophysical, geochemical and geodetical signals of
reawakening at Turrialba volcano (Costa Rica) after almost 150 years of quiescence. J. Volcanol. Geotherm. Res. 198, 416e432.
Moloney, T.P., Sims, K.W., Kaszuba, J.P., 2011. Uranium and Thorium Decay Series
Isotopic Constraints on the Source and Residence Time of Solutes in the Yellowstone Hydrothermal System, vol. 33. University of Wyoming National Park
Service Research Centre, pp. 187e207.
da, T., Szak
csy, I., 2008. Radon concentration measurements
Ne
acs, A., Cosma, C., Mo
in mofettes from Harghitaand Covasna Counties, Romania. J. Environ. Radioact.
99, 1819e1824.
da, T., Szaka
cs, A., Mo
 csy, I., Cosma, C., 2008. Radon concentration levels in dry
Ne
i, Romania, used for curative purposes.
CO2 emanations from Harghita Ba
J. Radioanal. Nucl. Chem. 277 (3), 685e691.
cs, A., Ne
da, T., Papp, Sz, Cosma, C., 2010. Soil radon and thoron
Papp, B., Szaka
studies near the mofettes at Harghita Bai (Romania) and their relation to the
eld location of fault zone. Geouids 10, 586e593.
rez, N.M., Herna
ndez, P.A., Padro
 n, E., Melia
n, G., Marrero, R., Padilla, G.,
Pe
Barrancos, J., Nolasco, D., 2007. Precursory subsurface 222Rn and 220Rn
degassing signatures of the 2004 Seismic Crisis at Tenerife, Canary Islands. Pure
Appl. Geophys. 164, 2431e2448.
Phuong, N.K., Harijoko, A., Itoi, R., Unoki, Y., 2012. Water geochemistry and soil gas
survey at Ungaran geothermal eld, central Java, Indonesia. J. Volcanol. Geotherm. Res. 229, 23e33.
Richon, P., Perrier, F., Sabroux, J.C., Trique, M., Ferry, C., Voisin, V., Pili, E., 2005.
Spatial and time variations of radon-222 concentration in the atmosphere of a
dead-end horizontal tunnel. J. Environ. Radioact. 78 (2), 179e198.
Saurer, M., Cherubini, P., Bonani, G., Siegwolf, R., 2003. Tracing carbon uptake from a
natural CO2 spring into tree rings: an isotope approach. Tree Physiol. 23 (14),
997e1004.
 , P., Kova
cs, T., 2011. Annual average
Somlai, J., Hakl, J., K
av
asi, N., Szeiler, G., Szabo
radon concentration in the show caves of Hungary. J. Radioanal. Nucl. Chem.
287 (2), 427e433.
Zimmer, M., Erzinger, J., 2003. Continuous H2O, CO2, 222Rn and temperature
measurements on Merapi Volcano, Indonesia. J. Volcanol. Geotherm. Res. 125
(1e2), 25e38.